[GitHub] [flink] pnowojski commented on pull request #20091: [FLINK-27570][runtime] Fix initialize base locations for checkpoint

[GitBox <gi...@apache.org>]

pnowojski commented on PR #20091:
URL: https://github.com/apache/flink/pull/20091#issuecomment-1174888262

   The initial set of ignored failure reasons was not well thought through and has been changed/adjusted over and over again, once we found a good example that a failure shouldn't be ignored. So:
   > Do you have an idea why FINALIZE_CHECKPOINT_FAILURE is not counted as a checkpoint failure? Looks strange to me as well.
   
   There is probably no good reason why.
   
   > If no strong reason, I go with @rkhachatryan , we should count FINALIZE_CHECKPOINT_FAILURE as well.
   
   👍 seems like a good change. Looking at the other failures `FINALIZE_CHECKPOINT_FAILURE` fits much better with things like:
   ```
               case IO_EXCEPTION:
               case CHECKPOINT_ASYNC_EXCEPTION:
               case CHECKPOINT_DECLINED:
               case CHECKPOINT_EXPIRED:
   ```
   Rather than checkpoint declined/subsumed. 
   
   I have similar doubts if maybe `TRIGGER_CHECKPOINT_FAILURE` shouldn't be ignored as well?


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org